Number of Electrons concerned in Metallic Conduction. 245 



of the experiments, this result indicates that there was 

 present no appreciable film or layer of any description ; and 

 consequently, for these liquids at least, the transition layer 

 which has been inferred from the phenomena of the reflexion 

 in air is not independent of the medium with which it is in 

 contact. In other words, the transition layer is to be re- 

 garded as a true region of interpenetration of two media ; 

 and the fact that it is appreciable in liquids and not in solids 

 must be attributed to the greater freedom of molecular 

 motions in the case of the former, and not to any change in 

 physical properties at the surface brought about by the force 

 of surface tension. 



The liquids chosen for this investigation include those for 

 which the previous observations have yielded values of n' less 

 than w , and one for which a value of n' greater than n has 

 been found. As the result of this work goes to show that 

 both of these sorts of discrepancy vanish when all possible 

 extraneous films are removed, it would seem a fair inference 

 that all of the discrepancies which have been found pre- 

 viously are to be attributed to the same causes, and that the 

 conclusion deduced from these experiments is a general one. 



Sheffield Scientific School of Yale 

 University, New Haven, Conn., 

 April 1911. 



XXVI. On the Number of Electrons concerned in Metallic 

 Conduction. By J. W. Nicholson, M.A., D.Sc* 



THE electron theory of the conduction of electricity 

 through metals, first developed by Drude, has in the 

 hands of later investigators, and notably of Riecke, Schuster, 

 J. J. Thomson, Lorentz, and Jeans, been placed on a firm 

 footing. But no certainty exists as to the number of free 

 electrons which must be supposed to be present in the atoms 

 of the various metals at ordinary temperatures, in order to 

 account for their optical properties, and in fact no detailed 

 investigation of the matter appears to have been made. It is 

 known that with certain hypotheses the mathematical results 

 are in fair agreement with experiment, but no agreement of 

 so exact a character as can be found in certain results of the 

 kinetic theory of gases, has been obtained. In this paper a 

 preliminary attempt is made to discriminate between certain 

 hypotheses with the aid of experimental results, and it is 

 shown that one hypothesis can be definitely selected as giving 



* Communicated bv the Author. 



